* Among women: 10 voted for rape proven beyond a reasonable doubt, 15 not proven.
* Among men: 17 voted for rape proven, 18 not proven.
I stressed to students, of course, that this was not a large sample, and most certainly not one representative of the country as a whole. But I think it ended up being a useful perspective for the students, in highlighting to everyone both (1) how close the division was, and (2) how little gender gap there was (with men actually being a little more likely than women to find rape proven beyond a reasonable doubt, but I'm not sure not to any statistically significant degree). My sense is that this was an important pedagogical tool, chiefly because it helped show people how people of their own age, sex, and social class can disagree on such matters (whether or not they conclude that such disagreement is indeed warranted).
The gender split is something I've noticed too when one of my own classes did a similar experiment -- albeit, not in our case, anonymously. In that case, the gender divide was even more pronounced -- I observed near unanimity among the men that the defendant should be found guilty of rape, while the women were approximately 50/50. Volokh's point still remains true, that people of one's own social position can nevertheless have widely divergent views on politically charged matters -- but I also think that this somewhat counter-intuitive gender division remains worthy of further examination as well.